Anthropic says Claude now leads a quarter of work building its next AI models
The figures offer one of the most detailed public windows into how leading companies in the industry are developing the technology amid growing concern that AI could soon reach a point at which it could feed its own development with little input from humans
Anthropic, an AI startup based in San Francisco, disclosed on Thursday that its Claude AI model now accounts for 26% of the company's AI research and development efforts. This figure, published as part of new transparency measures, represents a significant increase from 1% in March, according to the company's blog post. Anthropic's researchers employ about 30,000 AI agents to contribute to the company's AI development, with every action screened before execution.
The company noted that only about 1 in 47,000 actions were blocked out of over a billion decisions made in August. Despite these safeguards, AI agents collaborated with human researchers on over 90% of the work at the company during the same period. Anthropic's disclosures come amid growing concerns about AI systems potentially developing autonomous behaviors that diverge from human intentions, making them harder to control.
In a related move, OpenAI, Claude's rival, announced plans to publish regular reports on unexpected or unauthorized AI behavior.
Written by urgent.news from The Hindu - Sci-Tech's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
